A
Aravindh Mahendran
Researcher at Google
Publications - 23
Citations - 4016
Aravindh Mahendran is an academic researcher from Google. The author has contributed to research in topics: Convolutional neural network & Transfer of learning. The author has an hindex of 11, co-authored 23 publications receiving 3079 citations. Previous affiliations of Aravindh Mahendran include University of Oxford & Carnegie Mellon University.
Papers
More filters
Proceedings ArticleDOI
Understanding deep image representations by inverting them
TL;DR: In this article, a general framework was proposed to invert representations such as HOG and Bag of Visual Words (BOW) to reconstruct the image itself, which can be applied to CNNs too.
Posted Content
Understanding Deep Image Representations by Inverting Them
TL;DR: This paper conducts a direct analysis of the visual information contained in representations by asking the following question: given an encoding of an image, to which extent is it possible to reconstruct the image itself?
Journal ArticleDOI
Visualizing Deep Convolutional Neural Networks Using Natural Pre-images
TL;DR: This paper studies several landmark representations, both shallow and deep, by a number of complementary visualization techniques based on the concept of “natural pre-image”, and shows that several layers in CNNs retain photographically accurate information about the image, with different degrees of geometric and photometric invariance.
Proceedings Article
Object-Centric Learning with Slot Attention
Francesco Locatello,Dirk Weissenborn,Thomas Unterthiner,Aravindh Mahendran,Georg Heigold,Jakob Uszkoreit,Alexey Dosovitskiy,Thomas Kipf +7 more
TL;DR: An architectural component that interfaces with perceptual representations such as the output of a convolutional neural network and produces a set of task-dependent abstract representations which are exchangeable and can bind to any object in the input by specializing through a competitive procedure over multiple rounds of attention is presented.
Book ChapterDOI
Salient Deconvolutional Networks
TL;DR: A family of reversed networks is introduced that generalizes and relates deconvolution, backpropagation and network saliency, and is used to thoroughly investigate and compare these methods in terms of quality and meaning of the produced images, and of what architectural choices are important in determining these properties.